Mercury contamination of remote lakes has been attributed to increasing deposition of atmospheric mercury, yet historic deposition rates and inputs from terrestrial sources are essentially unknown. Sediments of seven headwater lakes in Minnesota and Wisconsin were used to reconstruct regional modern and preindustrial deposition rates of mercury. Whole-basin mercury fluxes, determined from lake-wide arrays of dated cores, indicate that the annual deposition of atmospheric mercury has increased from 3.7 to 12.5 micrograms per square meter since 1850 and that 25 percent of atmospheric mercury deposition to the terrestrial catchment is exported to the lake. The deposition increase is similar among sites, implying regional or global sources for the mercury entering these lakes.

We studied total mercury (THg) and methylmercury (MeHg) in eight streams, located in Oregon, Wisconsin, and Florida, that span large ranges in climate, landscape characteristics, atmospheric Hg deposition, and water chemistry. While atmospheric deposition was the source of Hg at each site, basin characteristics appeared to mediate this source by providing controls on methylation and fluvial THg and MeHg transport. Instantaneous concentrations of filtered total mercury (FTHg) and filtered methylmercury (FMeHg) exhibited strong positive correlations with both dissolved organic carbon (DOC) concentrations and streamflow for most streams, whereas mean FTHg and FMeHg concentrations were correlated with wetland density of the basins. For all streams combined, whole water concentrations (sum of filtered and particulate forms) of THg and MeHg correlated strongly with DOC and suspended sediment concentrations in the water column.

Mercury speciation, controls on methylmercury (MeHg) production, and bed sediment−pore water partitioning of total Hg (THg) and MeHg were examined in bed sediment from eight geochemically diverse streams where atmospheric deposition was the predominant Hg input. Across all streams, sediment THg concentrations were best described as a combined function of sediment percent fines (%fines; particles < 63 μm) and organic content. MeHg concentrations were best described as a combined function of organic content and the activity of the Hg(II)-methylating microbial community and were comparable to MeHg concentrations in streams with Hg inputs from industrial and mining sources. Whole sediment tin-reducible inorganic reactive Hg (Hg(II)_R) was used as a proxy measure for the Hg(II) pool available for microbial methylation. In conjunction with radiotracer-derived rate constants of ²⁰³Hg(II) methylation, Hg(II)_R was used to calculate MeHg production potential rates and to explain the spatial variability in MeHg concentration. The %Hg(II)_R (of THg) was low (2.1 ± 5.7%) and was inversely related to both microbial sulfate reduction rates and sediment total reduced sulfur concentration. While sediment THg concentrations were higher in urban streams, %MeHg and %Hg(II)_R were higher in nonurban streams. Sediment pore water distribution coefficients (log K_d’s) for both THg and MeHg were inversely related to the log-transformed ratio of pore water dissolved organic carbon (DOC) to bed sediment %fines. The stream with the highest drainage basin wetland density also had the highest pore water DOC concentration and the lowest log K_d’s for both THg and MeHg. No significant relationship existed between overlying water MeHg concentrations and those in bed sediment or pore water, suggesting upstream sources of MeHg production may be more important than local streambed production as a driver of water column MeHg concentration in drainage basins that receive Hg inputs primarily from atmospheric sources.